Benzophenone dicarboxylic acid antagonists of leukotriene B4. 2

Benzophenone dicarboxylic acid antagonists of leukotriene B4. 2. ... Modulators of Leukotriene Biosynthesis and Receptor Activation. Clint D. W. Brook...
1 downloads 0 Views 987KB Size
J. Med. Chem. 1990,33, 2807-2813 lysis. Cell preparations were 190% neutrophils and 290% viable. The effectiveness of compounds to inhibit binding of [3H]LTB4 to neutrophils was measured by using an adaptation of a radioligand-binding assay developed by Goetzl and Goldman.20The following were added to microcentrifuge tubes: 10 pL of DMSO containing different amounts of compound, 20 pL of radioligand (2.65 nM [SH]LTB4),and 500 pL of cells suspended at a concentration of 2 x 10’ cells/mL in Hanks’ balanced salt solution containing 0.1% ovalbumin. The tubes were then incubated at 4 O C for 10 min. After the incubation, 300 pL of a mixture of dibutyl and dinonyl phthalate ( 7 2 ) were added, and the tubes were centrifuged for 2 min. The liquid was then decanted and the bottom tip of the tube was cut off with a razor blade and placed in a counting vial. The radioactivity bound to the cell pellet was measured by scintillationspectrometry. Nonspecific binding was determined by measuring the amount of the label bound when cells and [3H]LTB4were incubated with a >2000-fold excess of nonradioactive ligand. Appropriate corrections for nonspecific binding were made when analyzing the data. Results are expressed as percent inhibition of specific [3H]LT134binding at the indicated

2807

concentrations. Each value is the mean of at least three replicates. The inhibitory activity of most compounds was evaluated on only one cell preparation. However an estimate of the precision of the measurements can be obtained from the inhibition observed with a reference compound, 5-[4-acetyl-5-hydroxyl-2-(2-propenyl)phenoxy]pentanenitrile, on all 102 cell preparations studied. At M, the mean percent inhibition and standard deviation for the reference compound were 93.9 and 3.9, respectively. At 10“ M, the correspondingvalues were 56.9 and 6.9. Assuming a linear correlation between percent inhibition and standard deviation, the following estimates were calculated for the precision at different percentages of inhibition: 90 4.2,80 f 5.0,60 6.6,40 f 8.2, 20 f 9.9, and 10 f 10.7. In a few cases where compounds were tested on more than one cell preparation, the precision of the measurements were equal to or better than these estimates (Le. compound 2, n = 3,73 2% at M, 27 5% at 10” M; compound 3, n = 4,95 f 0.5% at M, 60 2% at 10” M).

*

*

*

A c k n o w l e d g m e n t . We thank Mr. Mark Tomich for helping t o prepare some of t h e compounds in this paper.

Benzophenone Dicarboxylic Acid Antagonists of Leukotriene B4.2. Structure-Activity Relationships of the Lipophilic Side Chaint D. Mark Gapinski,* Barbara E. Mallett, Larry L. Froelich, and William T. Jackson Lilly Research Laboratories, Eli Lilly and Company, Indianapolis, Indiana 46285. Received September 18, 1989

A series of lipophilic benzophenone dicarboxylic acid derivatives were found to inhibit the binding of the potent chemotaxin leukotriene B, (LTB,) to ita receptor on intact human neutrophils. Activity at the LTB4receptor was determined by using a [3H]LTB4-bindingassay. The structure-activity relationship for the lipophilic side chain was systematically investigated. Compounds with n-alkyl side chains of varying lengths were prepared and tested. Best inhibition of [3H]LTB4binding was observed with the n-decyl derivative. Analogues with alkyl chains terminated with an aromatic ring showed improved activity. The 6-phenylhexyl side chain was optimal. Substitution on the terminal aromatic ring was also evaluated. Methoxyl, methylsulfinyl, and methyl substituents greatly enhanced the activity of the compound. For a given substituent, the para isomer had the best activity. Thus the nature of the lipophilic side chain can greatly influence the ability of the compounds to inhibit the binding of LTB4 to its receptor on intact human neutrophils. The most active compound from this series, 84 (LY223982),bound to the LTB, receptor with an affinity approaching that of the agonist. During the course of our search for compounds which blocked the biological effects of leukotriene B, (LTB,), a series of benzophenone dicarboxylic acid derivatives with lipophilic side chains was synthesized and found to inhibit the binding of LTB, t o its receptor(s) on intact human neutrophils. A structureactivity relationship was studied for these compounds t o determine which structural elements were required for activity and t o maximize the activity of this series at the LTB, receptor. This paper will discuss a portion of this structureactivity study involving structural modifications of the lipophilic portion of these antagonists. T h e preceeding paper discussed the structure-activity relationships for t h e benzophenone dicarboxylic acid portion of these compounds.’ Chemistry T h e alkoxybenzophenone diacids for this study were prepared as illustrated in Scheme I. Alkylation of the key hydroxybenzophenone diester 1-3 with the halide or mesylate of t h e desired side chain yielded alkoxybenzophenone diesters 33-66. Basic hydrolysis then yielded the target compounds 67-101. T h e hydroxybenzophenone diester was synthesized by using a three-step procedure beginning with a Friedel-Crafts reaction using (3-carbThis work was presented in part at the 196th Meeting of the American Chemical Society, Los Angeles CA, September, 1988.

ethoxy)benzoyl chloride and ethyl 3-(2-methoxyphenyl)propanoate, yielding 1-1. Dealkylation with pyridine hydrochloride gave the corresponding hydroxy diacid 1-2, which was esterified, giving 1-3. T h e methanesulfonate alkylating agents were prepared directly from their corresponding alcohols with methanesulfonyl chloride and triethylamine in ether as shown in Scheme 11. T h e alcohols (Table 11) unless otherwise indicated were prepared from the corresponding carboxylic acids (Table I) via lithium aluminum hydride reduction. Unsaturated carboxylic acid precursors (Table I) were generally prepared by using Wittig olefination chemistry. T h u s the ylide derived from (4-carboxybuty1)triphenylphosphonium bromide was reacted with the desired aldehyde, giving t h e (E)-styrene derivative as the major product.2 I n no case could useful quantities of the (2)styrene derivative be isolated. Acetylene derivative 26 was prepared by palladium-catalyzed coupling of 4-bromoanisole and 5-hexyne-1-01 (Scheme IIU3 Compounds with saturated aryl side chains were prepared from either un(I) Gapinski, D. M.; Mallett, B. E.; Frohlich, L. L.; Jackson, W. T. J. Med. Chem., Preceding paper in this issue. (2) Maryanoff, B. E.; Reitz, A. B.; Duhl-Emswiler, B. A. J . Am. Chem. SOC.1985, 107, 217-226. (3) Takahashi, S.; Kuroyama, Y.;Sonogashira, K.; Hagihara, N. Synthesis 1980,627.

0022-2623/90/1833-2807$02.50/0 0 1990 American Chemical Society

Gapinski et al.

2808 Journal of Medicinal Chemistry, 1990, Vol. 33, No. 10 Table I. Data for Intermediate w-Aryl Acids and Esters

R(CH,),COOR no.

routeC

R' 90 yield mp, OC anal. Et 33 a C, H 1 PhO Et 47 a C, H 2 PhS H 64 a C, H 3 (E)-PhCH=CH c H 95 a C, H 4 Ph(CHz)z 28 76-78 C, H 5 (E)-(4-MeO)PhCH=CH B H 37 83-84 C, H 6 (E)-(4-MeS)PhCH=CH B H 55 a C, H 7 (E)-(3-MeO)PhCH=CH B H 18 83-84 b 8 (E)-(2-MeO)PhCH=CH E3 H 48 62-64 C, H 9 (E)-(2,4-(MeO),)PhCH=CH B H C H 89 a b 10 (4-MeO)Ph(CHZ), C H 84 a b 11 (4-C1)Ph(CHz)Z 12 (E)-(4-Cl)PhCH=CH B H 32 63-65 C, H c H 79 a b 13 (4-F)Ph(CHz), B H 10 a b 14 (E)-(4-F)PhCH=CH 38 60-61 C, H 15 (E)-(4-Me)PhCH=CH B H 3 a b 16 (E)-(4-OH)PhCH=CH B H OCompound obtained as an oil. *Analysis not obtained. Spectral data consistent with indicated structure. C A prepared by alkylation of bromo ester. B: prepared by Wittig olefination. C: prepared by Wittig olefination followed by catalytic hydrogenation. R

A A B

Scheme I

Table 11. Data for Intermediate w-Aryl Alcohols R(CH&,OH

,COOEt

?P, C anal. 17 PhO a C, H a b 18 PhS 19 (E)-PhCH=CH a C, H 20 Ph(CH2)z a C,H 21 (E)-(4-MeO)PhCH=CH 65-66 C, H 22 (E)-(4-MeS)PhCH=CH A 71 68-70 C, H 23 (E)-(3-MeO)PhCH=CH A 81 a C, H 24 (E)-(2-MeO)PhCH=CH A 57 a C, H 25 (E)-(2,4-(MeO),)PhCH=CH A 80 a b 26 (4-MeO)PhC=C C 3 7 a C,H 27 (4-MeO)Ph(CH2), B 57 a C, H 28 (CCl)Ph(CHZ)z B 88 a C, H b 29 (4-F)P h(CHJ2 B 7 5 a 30 (E)-(4-Me)PhCH=CH A 43 a C, H 31 (4-MeCO)Ph(CH2)2 D 43 45-46 C, H 32 (E)-(4-OH)PhCH=CH A 40 a C, H "Compound obtained as an oil. *Analysis not obtained. Spectral data consistent with indicated structure. 'A: prepared by LAH reduction of ester. B: prepared by catalytic hydrogenation of ester followed by LAH reduction. C: prepared by arene/ acetylene coupling. D: prepared by acylation of 6-phenylhexanol. no.

/COOEt

0 EtOOC

PyrHCi. 180 "2 9

1-1

%

R

routec yield A 86 A 2 4 A 77 B 9 1 A 80

"ocMoH 0

(COOH

EtOH, H +

1-2 0

/COOEI

EtoocMo HoocMo, 1. RX. K2C03

9

2. KOH, MeOHIH20

1-3

0

/COOH

Scheme I1

+ P h 3 P ~ C O O -

1 4

saturated alcohols or acids by catalytic hydrogenation using P d on carbon as the catalyst. Derivatives 1 and 2 were prepared by alkylation of ethyl 4-bromovalerate with sodium phenoxide and thiophenoxide, respectively. 4-Acetyl derivative 31 was prepared from 20 by direct FriedelCrafts acylation with acetyl chloride. Sulfoxide and sulfone derivatives 47, 48 and 52, 53 (Table 111) were prepared from sulfides 46 and 51 by oxidation with either 1or 2 equiv o€ mCPBA in methylene chloride. (2)-Styrene 91 was prepared by catalytic hydrogenation of acetylene derivative 92, using a Lindlar catalyst.

1LiAIH.

-

OH CHsS02CI

R

-

p

TEA, Et20

OSO,CH,

R-3"""

Results The structureactivity relationships for this series were studied by evaluating the ability of compounds t o inhibit the binding of [3H]LTB4to its receptor($ on intact human polymorphonuclear leukocytes. The nature of the lipophilic side chain had a profound influence of the ability

of compounds to bind to leukotriene B4 receptors. Compounds 67-72 demonstate t h e effect of chain length on activity. In this series of n-alkoxy derivatives, optimal activity was observed when the length of the chain was 10 carbons (71). Longer and shorter chain lengths gave

Antagonists of Leukotriene Bq. 2

Journal of Medicinal Chemistry, 1990, Vol. 33, No. 10 2809

Scheme 111

erably less active than t h e related (E)-olefin. Linear acetylene derivative 92 inhibited to a degree intermediate to that of the two olefin isomers. Replacement of the carbon-carbon double bond by heteroatoms, as in 79 and 80, or polar groups, as in 81 and 82, gave compounds of dramatically reduced activity. The effect of various substituents on t h e terminal aromatic ring of (w-phenylhexy1)benzophenone diacids was investigated. Substitution of the aromatic ring gave compounds of equal or greater activity in t h e binding assay than the unsubstituted phenyl derivatives. The addition of a methoxyl substituent to the terminal phenyl ring yielded two of the most potent compounds in this series, 84 and 93. Methyl (98) and methylsulfinyl (86) substituents also gave very potent inhibitors. Acetyl derivative 99 showed slightly less activity when compared to t h e isoelectronic sulfoxide 86. Halogen (96 and 97) and methylmercapto (85) substituents gave compounds with activity no greater than the unsubstituted compound (77). The effect of the position of the substituent on t h e terminal phenyl ring was probed with the methoxyl-substituted derivatives 84,88 and 89 and with compounds 93-95. there was not much difference in the activity of the isomers although the para-substituted methoxyl derivatives tended to be more inhibitory. Compound 90, having methoxyl groups in both t h e ortho and para positions, showed slightly less activity than either monomethoxyl derivative. Hydroxyl-substituted compound 100 was comparable in activity to the 4-chloro and 4-acetyl derivatives.

Table 111. Data for Benzophenone Diester Intermediates 0

/COOEI

Etooc-dcd,, no. R 70yield" anal. nn I& 33 CH,' 71 34 72 35 64 36 37 83 72 38 52 39 25 40 67 41 33 42 45 43 7 44 59 45 42 46 44 47 73 48 (CH&SOZPh 63 49 (E)-PhCH=CH(CHz), 66 50 (E)-(4-MeO)PhCH=CH(CHA 70 51 73 52 68 53 41 54 18 55 20 56 55 57 66 58 94 59 60 93 47 61 56 62 27 63 64 (4-MeCO)Ph(CHz)6 33 65 (E)-(4-CH3SOzO)PhCH=CH(CH,), 42 30 66 (Eb(2-thien~l)CH=CH(CH,)~ All compounds obtained as oils unless otherwise indicated. *Analysis not obtained. Spectral data consistent with indicated structure. 'Mp: 54-56 "C. dMp: 70-72 "C. ~~

compounds of lesser activity. Alkyl chains such as the decyloxy side chain of 71 are frequently targets for extensive oxidative metabolic degradation. T o retard such metabolism, derivatives were prepared in which the alkyl chain was terminated with an aromatic ring. T h e w-phenyl derivatives were found to be more effective a t inhibiting specific [3H]LTB4binding than t h e corresponding n-alkyl derivatives. A similar dependence of activity on chain length was seen with w-phenyl derivatives 73-78. Optimal inhibition was observed with t h e n-hexyl chain terminated with a phenyl group (77). Longer and shorter chain derivatives were less active. A terminal 2-thienyl group, as in 101, may substitute for the terminal phenyl group with no decrease in activity. A derivative with a trans carbon-carbon double bond in the chain, such as 83, possesses activity similar to that of the corresponding saturated chain compound (77). The effect of olefin geometry on activity was investigated by comparing compounds 84 and 91. T h e (2)-olefin was consid-

Discussion The alkoxybenzophenone derivatives in Table IV inhibited the binding of leukotreine B, to its receptor(s) on intact human neutrophils a t concentrations ranging from to 10" M (Table IV). The most active compounds in this series inhibited 50% or more a t lo-' M. The length of the lipophilic tail profoundly effected the degree of inhibition. Compounds with either nine methylenes and a methyl group or six methylenes and an aromatic ring inhibited the most. The approximate length of these derivatives, from the aromatic carboxylic acid t o the end of the alkyl or phenalkyl tail, is similar to t h e length from C1 to C20 of LTB,. The most potent derivatives had an alkoxy tail terminated by a substituted aromatic ring. The electronic properties of substituents on the terminal aromatic ring did not influence binding affinity. Compounds with electron-withdrawing (methylsulfinyl) and donating (methoxyl) groups had similar activity. T h e most active alkoxybenzophenone synthesized was compound 84 (LY223982). It, therefore, was selected for more extensive study. T h e IC50for inhibition of [3H]LTB4binding was only 7-fold higher than t h e value obtained for nonradioactive LTB, (Table V). Hence, 84 displaces [3H]LTB4 nearly as well as t h e natural ligand. Pharmacological studies indicate that 84 is a selective antagonist of LTB, under both in vitro4 and in vivo5 conditions. In conclusion, t h e alkoxybenzophenones represent a class of potent and selective antagonists of leukotriene B,. Modifications of the lipid side chain of these antagonists has produced compounds of considerably enhanced affmity for the LTB, receptor. The most potent members of this series bind almost as well as LTB,. These antagonists may serve as useful agents in elucidating the pathophysiological (4) Jackson, W. T.; Boyd, R. J.; Froelich, L. L.; Goodson, T.; Bollinger, N. G.; Herron, D. K.; Mallett, B. E.; Gapinski, D. M. FASEB J. 1988,2, A1110. (5) Jackson, W. T.; Froelich, L. L.; Goodson, T.; Herron, D. K.; Mallett, B. E.; Gapinski, D. M. Pharmacologist 1988,30, A206.

Gapinski et al.

2810 Journal of Medicinal Chemistry, 1990, Vol. 33, No. 10 Table IV. Data for Benzophenone Dicarboxylic Acid LTB4 Receptor Antagonists n

no. 67 68 69 70 71 72 73

R O

CH,

% yield

72

75 11 77

79 56 77

mp, OC 225-227 226-227 157-160 144-146 115-117 114-1 16 177-178 147-149 188-189 116-117 99-101 116-118 116-118 165-166 135-137 197-199 125-1 28 151-152 138-141 119-1 2 1 152-153 122-125 132-136 108-110

anal. C. H

75 74 75 58 76 62 77 23 78 47 79 66 80 17 81 93 82 70 83 64 84 74 85 56 86 52 87 45 88 33 89 72 90 13 91 144-146 59 92 100-102 54 93 76 88-90 94 125-127 74 95 119-1 2 1 75 96 118-120 51 97 (4-F)Ph(CH2j," 148-150 57 (E)-(4-Me)PhCH=CH(CH2), 98 139-141 66 (4-MeCO)Ph(CHdc 99 166-169 37 100 141-143 25 101 nEstimates of the precision at different inhibition percentages are 90 f 4.2, 80 Elemental analysis not obtained. Spectral data consistent with assigned structure. 74

Table V. Inhibition of Specific Binding of [,H]-LTB4to Human Neutronhils comDound no nM LTB, 5 1.9 f 0.05 LY223982 5 13.2 f 2.2 Number of experiments conducted. bValues are expressed as mean f standard error of the mean. role of LTB, in human disease. One of these antagonists, LY223982,has been selected for clinical trials in man.

Experimental Section Melting points were determined using a Thomas-Hoover capillary melting point apparatus and are uncorrected. N M R spectra were obtained with a GE QE-300 spectrometer. The following abbreviations are used to denote signal patterns: s = singlet, d = doublet, t = triplet, br = broad, m = multiplet. All chemical shifts are reported relative to a tetramethylsilane internal standard. IR spectra were determined with a Nicolet DXlO FT-IR spectrometer. Mass spectral data were determined with a CEC-21-110 electron-impact mass spectrometer. All spectroscopic and analytical data were determined by the Physical Chemistry Department (MC525) of the Lilly Research Laboratories. Vapor-phase chromatography (VPC) was carried out with a Hewlett-Packard gas chromatograph equipped with a Hewlett-Packard 3392A integrating recorder and an H / P methyl silicon capillary column. T H F which had been stored over 4A molecular sieves

percent inhibn of specific [,HI LTB, bindineP M 10" M M 10" M 0 7 0

10 35 60 29 3 3 19

0 0 0

0

0 0

7 0

0

36 28

14

34

3 40

21

85 87 29 16

0

0 4

92 95 79 95 89 91 100 96 76 90 96 89 91 93

88

18 86 60 49 88 66 31 45 68 46 3 42

77

22

96 90 90

73 46 43

36 31 13 25 20 0

9 18 5 15 6 24 11

6

21 f

5.0, 60 f 6.6, 40

f

8.2, 20 f 9.9, and 10 f 10.7.

was further dried by distillation from sodium/benzophenone ketyl immediately prior to use. E t h y l 5-[3-(Ethoxycarbonyl)benzoyl]-2-methoxybenzenepropanoate (1-1). A solution of 3-carbethoxybenzoyl chloride (34.51 g, 0.163 mol) in methylene chloride (50 mL) was added to a suspension of aluminum chloride (64.95 g, 0.488 mol) in methylene chloride (500 mL) a t 0 O C . After 1 h, ethyl 3-(2methoxypheny1)propanoate (33.85 g, 0.163 mol) was dissolved in methylene chloride (50 mL) and added dropwise. The reaction mixture was stirred at 0 "C for 4 h and then poured into a mixture of ice and HC1 (concentrated). The mixture was swirled until all of the aluminum salh had dissolved (ca. 1 h). The layers were separated, and the organic layer was washed with a saturated sodium bicarbonate solution, dried over MgS04, and evaporated. The crude product was purified by chromatography over silica gel using a Waters Prep LC-500. A 0-30% ethyl acetatelhexane gradient gave excellent separation of the reaction mixture. Appropriate pure fractions were combined and evaporated, p'ovid9 the desired benzophenone (53.6 g, 86%)as a pale yellow oil which crystallized on standing. Mp 54-56 OC. NMR (CDCl,): 6 8.41 (s, 1 H), 8.25 (d, J = 6.7 Hz, 1 H), 7.94 (d, J = 6.7 Hz, 1 H), 7.73 (m, 2 H), 7.58 (t,J = 7.8 Hz, 1H), 6.93 (d, J = 9.7 Hz, 1 H), 4.42 (q, J = 7.8 Hz, 2 H), 4.13 (q, J = 7.8 Hz, 2 H), 3.93 (s, 3 H), 3.02 (t, J = 9.7 Hz, 2 H), 2.64 (t,J = 9.7 Hz, 2 H), 1.43 (t, J = 6.7 Hz, 3 H), 1.24 (t, J = 6.7 Hz, 3 H). IR (CHCl,, cm-'): 1719, 1652, 1259. MS: m / e 384 (M+). Anal. (C22H2406):C, H. 5-(Carboxybenzoyl)-2-hydroxybenzenepropanoicAcid (1-2). Ethyl 5-[3-(ethoxycarbonyl)benzoyl]-2-methoxybenzene-

Antagonists of Leukotriene B,. 2

Journal of Medicinal Chemistry, 1990, Vol. 33, No. 10 2811

propanoate (41.5 g, 0.108 mol) and pyridine hydrochloride (410 g, 3.55 mol) were mixed together and heated to 180 "C for 4 h. The mixture was then allowed to cool. When the reaction mixture had cooled to 100 OC, an equal volume of water was added. As the mixture cooled further, the title product precipitated from solution. The product was filtered, washed thoroughly with water, and dried at 80 OC in a vacuum dessicator. The material prepared in this manner (31.1, 92%) was sufficiently pure for further transformations. An analytically pure sample could be prepared by recrystallization from ethanol/water. Mp: 197-200 "C. NMR (DMSO): 6 12.65 (br s, 2 H), 10.55 (br s, 1H), 8.17 (m, 2 H), 7.90 (d, J = 9.0 Hz, 1 H), 7.67 (t, J = 7.8 Hz, 1 H), 7.62 (br s, 1 H), 7.52 (d, J = 9.0 Hz, 1 H), 6.94 (d, J = 9.0 Hz, 1H), 2.82 (t, J = 7.2 Hz, 2H), 2.52 (t,J = 7.2 Hz, 2 H). IR (KBr, cm-'): 3200 (br), 1692, 1728. MS: m / e Mt + 18. Anal. (C17H1,06): C, H. E t h y l 5- [3-( E t h o x y c a r b o n y l )benzoyl]-2-hydroxybenzenepropanoate (1-3). Diacid 1-2 (31.1 g, 0.099 mol) was suspended in absolute ethanol (600 mL). Sulfuric acid (1 mL) was added as catalyst. The mixture was heated a t reflux for 4 days and then cooled to 25 "C. After concentrating in vacuo, ethyl acetate was added to the residue and the solution was washed with water, dried over sodium sulfate, and evaporated. Purification of the crude mixture was performed by column chromatography (Waters Prep LC-500) using a 20-50% ethyl acetate/ hexane gradient eluent. Combination and evaporation of the pure fractions gave a white solid which crystallized on standing. Recrystallization from ethyl acetate and hexane gave the desired product (21.52 g, 56%). Mp: 68-70 "C. NMR (CDClJ: 6 8.38 (br s, 1 H), 8.25 (m, 2 H), 7.44 (s, 1 HI, 7.63-7.53 (m, 2 H), 6.95 (d, J = 8.4 Hz, 1 H), 4.41 (q, J = 6.0 Hz, 2 H), 4.18 (q, J = 6.0 Hz, 2 H), 2.97 (d, J = 6.0 Hz, 2 H), 2.77 (d, J = 6.0 Hz, 2 H), 1.42 (t,J = 7.2 Hz, 2 H), 1.27 (t,J = 7.2 Hz, 3 H). IR (CHC13,cm-'): 3300 (br), 1716, 1232. MS: m / e 370 (M'). Anal. (C21H2206): C, H. Procedure for Alkylation of the Benzophenone Nucleus. E t h y l 2-[(6-Phenylhexyl)oxy]-5-[3-(ethoxycarbonyl)benzoyl]benzenepropanoate (43). To a solution of ethyl 5[3-(ethoxycarbonyl)benzoyl]-2-hydroxybenzenepropanoate(2.89 g, 7.81 mmol) in DMF (50 mL) was added sodium hydride (360 mg, 60% oil dispersion, 8.64 mmol) in small portions. After initial gas evolution had subsided, the mixture was stirred under nitrogen a t 25 OC for 30 min. The methanesulfonate ester of 6-phenylhexanol(2.00 g, 7.81 mmol), dissolved in a small quantity of DMF, was added and the mixture was warmed to 65 "C. After stirring overnight, the mixture was cooled and carefully poured into ice water. The resulting mixture was extracted three times with ethyl acetate. The combined extracts were washed with brine and dried over MgSOI. After evaporation of the solvent, the crude product was purified by column chromatography (Waters Prep LC-500) using a 0-20% ethyl acetate/hexane gradient as the eluent. Pure fractions were combined to yield the title compound (1.88g, 45%) as a pale yellow oil. NMR (CDClJ: 8 8.43 (s, 1H), 8.27 (d, J = 9.7 Hz, 1 H), 7.96 (d, J = 9.7 Hz, 1 H), 7.76 (m, 2 H), 7.59 (t, J = 6.8 Hz, 1 H), 7.31 (t, J = 5.8 Hz, 3 H), 7.20 (d, J = 5.8 Hz, 2 H), 6.91 (d, J = 9.7 Hz, 1 H), 4.44 (4,J = 7.8 Hz, 2 H), 4.03 (m, 4 H), 3.03 (t, J = 5.8 Hz, 2 H), 2.66 (m, 4 H), 1.93-1.48 (m, 8 H), 1.44 (t, J = 6.8 Hz, 3 H), 1.25 (t, J = 6.8 Hz, 3 H). IR (CHCl,, cm-'): 1718,1600,1258. MS: m / e 530 (Mt). Anal. (C3H306): C, H. General Procedure for Ester Hydrolysis: 2-[ (6-Phenylhexyl)oxy]-5-(3-~arboxybenzoyl)benzenepropanoicAcid (77). Ethyl 2-[(6-phenylhexyl)oxy]-5-[3-(ethoxycarbonyl)benzoyl]benzenepropanoate (1.88 g, 3.52 mmol) was dissolved in a 10:1 mixture of ethanol and water (100 mL). KOH (0.79 g, 14.1mmol) was added in one portion. The mixture was stirred for 1h at 25 "C. After this time TLC indicated that all of the starting material was consumed. The mixture was then concentrated in vacuo and the residue was partitioned between water and ether. The layers were separated, and the aqueous layer was acidified to pH = 2 with 1N HCl. Ethyl acetate was added to dissolve the precipitate. The aqueous layer was extracted two additional times with ether. The combined organic extracts were dried and evaporated, giving a white solid. The desired diacid was further purified by recrystallization from ethyl acetate/hexane yielding pure 77 (1.04 g, 62%). MP: 99-101 "C. NMR (CDC13 + DMSO-d6): 6 8.33 (s, 1 H), 8.27 (d, J = 9.7 Hz, 1 H), 8.15 (d, J = 9.7 Hz, 1 H), 7.88

(d, J = 8.7 Hz, 2 HI, 7.66 (m, 2 HI, 7.30 (m, 3 H), 7.22 (d, J = 8.7 Hz, 2 H), 6.97 (d, J = 9.7 Hz, 1 H), 4.10 (t, J = 7.8 Hz, 2 H), 2.99 (t, J = 4.8 Hz, 2 H), 2.75 (d, J = 4.8 Hz, 2 H), 2.66 (d, J = 7.8 Hz, 2 H), 1.94-1.43 (m, 8 H). IR (CHC13,cm-'): 3028, 1707, 1652. MS: m / e 474. Anal. (CmHYO6): C, H. General Procedure for Sulfoxide Formation: Ethyl 5[3-(Ethoxycarbonyl)benzoyl]-2-[4-(phenylsulfiny1)butoxylbenzenepropanoate (47). Ethyl 5-[3-(ethoxycarbonyl)benzoyl]-2-[4-(phenylthio)butoxy]benzenepropanoate (317 mg, 0.59 mmol) was dissolved in methylene chloride (25 mL) and cooled to -78 "C. With good stirring, m-chloroperbenzoic acid (127 mg, 0.59 mmol) was added to the reaction mixture in one portion. After stirring for 5 min, the external cooling bath was removed and the reaction was allowed to warm slowly for 10 min. Several drops of dimethyl sulfide were added to consume any excess oxidant. Ethyl acetate was then added and the mixture was washed with sodium bicarbonate solution. The reaction mixture was dried over MgSO, and concentrated in vacuo. The residue was purified by preparative thin-layer chromatography, providing the title compound as a colorless oil (143 mg, 44%). NMR (CDClJ: 6 8.40 (s, 1 H), 8.24 (d, J = 7.8 Hz, 1 H), 7.92 (d, J = 7.8 Hz, 1 H), 7.72-7.42 (m, 8 H), 4.39 (4, J = 5.8 Hz, 2 H), 4.12 (m, 4 H), 2.93 (m, 4 H), 2.60 (t,J = 5.8 Hz, 2 H), 2.17-1.80 (m, 4 H), 1.42 (t, J = 5.8 Hz, 3 H), 1.25 (t, J = 5.8 Hz, 3 H). IR (CHC13, cm-'): 3018, 1720, 1287. MS: m / e 550 (Mt). Anal. (C31Ha0,S): C, H. General Procedure for Sulfone Formation: Ethyl 5-[3(Et hoxycarbonyl)benzoyl]-2-[4 4 phenylsulfonyl)butoxy]benzenepropanoate (48). Ethyl 5-[3-(ethoxycarbonyl)benzoyl]-2-[4-(pheny1thio)butoxylbenzenepropanoate (326 mg, 0.62 mmol) was dissolved in methylene chloride (25 mL). mChloroperbenzoic acid (262 mg, 1.22 mmol) was added in one portion a t 25 OC. The mixture was stirred a t that temperature for 2 h. Several drops of dimethyl sulfide were added to consume any residual oxidant. Ethyl acetate was then added and the mixture was washed with sodium bicarbonate solution. The reaction mixture was dried over MgSO, and concentrated in vacuo. The residue was purified by preparative thin-layer chromatography, providing the title compound as a colorless oil (252 mg, 73%). NMR (CDCl3): 6 8.37 (9, 1 H), 8.23 (d, J = 6.7 Hz, 1 H), 7.91 ( J = 6.8 Hz, 2 H), 7.80-7.51 (m, 7 H), 6.84 (d, J = 8.7 Hz, 1 H), 4.39 (9, J = 6.8 Hz, 2 H), 4.15-4.00 (m, 4 H), 3.12 (br t, 2 H), 2.92 (t, J = 5.8 Hz, 2 He, 2.54 (t, J = 5.8 Hz, 2 H), 1.98 (br m, 4 H), 1.40 (t, J = 5.8 Hz, 3 H), 1.23 (t, J = 5.8 Hz, 3 H). IR (CHC13, cm-'): 3018, 1720, 1288. MS: m / e 566 (M'). Anal. (C31Hg08S): C, H. Ethyl 4-(Pheny1thio)butyrate (2). NaH (3.89 g, 97 mmol) was added portionwise to a DMF solution (250 mL) of thiophenol (10 mL, 97 mmol). When addition was complete, the mixture was stirred for 1 h a t 25 "C. Ethyl 4-bromobutyrate (10.8 g, 106 mmol) was added dropwise and the reaction mixture was warmed to 65 "C. After stirring overnight at this temperature, the mixture was cooled and poured into an ethyl acetate/water mixture. The organic layer was washed again with water and then with brine. After drying over MgSO, and evaporation of the solvent, the residue was purified by chromatography. A 0 4 % ethyl acetate/hexane gradient elution system gave good results. Combination of pure fractions gave the desired compound (10.18 g, 47%) as a pale yellow oil. NMR (CDC13): 6 7.28-7.16 (m, 5 H), 4.15 (4,J = 7.8 Hz, 2 H), 3.02 (t, J = 5.8 Hz, 2 H), 2.50 (t,J = 5.8 Hz, 2 H), 2.02 (t,J = 5.8 Hz, 2 H), 1.29 (t, J = 5.8 Hz, 3 H). IR (CHC13, cm-'): 3000, 1726. MS: m / e 224 (M'). Anal. (Ci&i602S): C, H. Ethyl 4-phenoxybutyrate (1) was prepared by using the same procedure as for 2 with phenol being substituted for thiophenol (31%). NMR (CDCI,): 6 7.37-7.24 (m, 2 H), 7.02-6.86 (m, 3 H), 4.16 (4,J = 6.8 Hz, 2 H), 4.03 (t,J = 6.0 Hz, 2 H), 2.57 (t, J = 6.0 Hz, 2 H), 2.16 (t, J = 6.0 Hz, 2 H), 1.29 (t,J = 6.8 Hz, 3 H). IR (CDCl,, cm-'): 2978, 1728. Anal. (C12H1803): C, H. 6-(4-Acetylphenyl)hexanol(31). 6-Phenylhexanol(5.34 g, 30 mmol) dissolved in methylene chloride (25 mL) was added dropwise to a suspension of AlC13(39.9 g, 300 mmol) in methylene chloride (250 mL) at 0 "C. Acetyl chloride (4.26 mL, 600 mmol) was then added dropwise and the mixture was allowed to warm to 25 OC. The reaction was stirred overnight and then poured into a mixture of ice and HCl (concentrated). The mixture was

2812 Journal of Medicinal Chemistry, 1990, Vol. 33, No. 10 stirred until all of the aluminum sal& had dissolved (ca.1h). The layers were separated. The organic layer was washed with aqueous bicarbonate, dried over MgS04, and evaporated in vacuo. The mixture was purified by chromatography (Waters Prep LC-500) using a 0-50% ethyl acetate/hexane gradient. Pure fractions were combined to yield alcohol 31 (2.85g, 43%) as a waxy solid. Mp: 45-46 "C. NMR (CDCl3): 8 7.90 (d, 6.8Hz, 2 H), 7.28 (d, 6.8Hz, 2 H), 3.64 (t, J = 5.8Hz, 2 H), 2.69 (t, J = 5.8 Hz, 2 H), 2.60 (2, 3 H), 1.73-1.33 (m, 8 H). IR (CHC13,cm-I): 2860,1678. MS: m / e 220 (M+). Anal. (C14HNO2): C, H. General Procedure for Carboxylic Acid Reduction Using LiAlH, LiAlH4: (E)-6-(4-Methoxyphenyl)hexd-en-l-ol(21). (16.01 g, 0.459 mol) was suspended in ether (1.0 L) and cooled to 0 "C. (E)-6-(4-methoxyphenyl)hex-5-enoic acid (25.26 g, 0.115 mol) dissolved in ether (100 mL) was added dropwise. The mixture was stirred at 25 "C for 4 h after the addition was complete. The mixture was then cooled to 0 "C. Water (16.0mL) was carefully added dropwise followed by 3 M KOH (16.0mL) and more water (54.0 mL). The entire mixture was stirred for 1 h and then filtered. The filtrate was dried over MgSO, and evaporated in vacuo. The solid residue was purified by recrystallization from ethyl acetate and hexane. The desired alcohol (19.01g, 80%)was obtained as white platelets. NMR (DMSO-d,): 6 7.31 (d, J = 9.0 Hz, 2 H), 6.86 (d, J = 9.0 Hz, 2 H), 6.32 (d, J = 16.2 Hz, 1 H), 6.12 (m, 1 H), 3.74 (9, 3 H),3.42 (br m. 2 H), 2.15 (br q, 2 H), 1.43 (br m, 4 H). IR (CHCl,, cm-'): 3625,1608, 1511. MS: m/e 206 (M'). Anal. (cl3H1&): C, H. General Procedure for Wittig Olefination: (E)-6-(4Met hoxypheny1)pent-5-enoic Acid (5). Hexamethyldisilazane (100 mL, 0.434 mol) was dissolved in dry THF (500 mL) and cooled to 0 "C under a nitrogen atmosphere. n-BuLi (250mL, 1.6 M solution in hexane) was added dropwise via cannula. The mixture was stirred for 15 min after completion of the addition. (4-Carboxybutyl)triphenylphosphoniumbromide (80.25g, 0.181 mol) was suspended in dry THF (1 L) under a nitrogen atmosphere. The lithium hexamethyldisiiazide solution was then added via cannula. A deep orange-red color developed after half of the base was added. The mixture was stirred at 25 OC for 1 h after addition of the base was finished. Anisaldehyde (24.6g, 0.181 mol) was added via syringe. The red color of the ylide was quenched almost immediately. The reaction mixture was stirred for an additional 2 h at 25 "C. Water (1L) was added carefully with stirring. The layers were separated, and the aqueous layer was washed with ether twice. The aqueous layer was then acidified to pH = 2 with concentrated sulfuric acid and extracted thoroughly with ethyl acetate. The combined extracts were dried over MgS04 and evaporated in vacuo. The residue was purified by crystallization from ethyl acetate/hexane giving pure 5 (11.17g, 28%) as white needles. MP: 76-78 "C. NMR (CDCI,): 6 7.32 (d, J = 9.7 Hz, 2 H), 6.87 (d, J = 9.7 Hz, 2 H), 6.39 (d, J = 15.5 Hz, 1 H), 7.17 (m, 1 H), 3.83 (s, 3 H), 2.43 (t, J = 6.8 Hz, 2 H), 2.28 (q, J = 7.8 Hz, 2 H), 1.85 (m, 2 H). IR (CHC13,em-'): 2955, 1709. MS: m/e 220 (M+). Anal. (C13H1,jOz): C, H. None of the corresponding cis-olefin isomer could be isolated from the mother liquor from the recrystallization. General Procedure for Double-Bond Reduction via Catalytic Hydrogenation: 6-Phenylhexanoic Acid (4). 6Phenylhex-5-enoic acid (6.8g, 35.8 mmol) was dissolved in ethyl acetate (100mL) in a fiberglass-coated hydrogenation flask. Pd/C (5%, ca. 100 mg) was added and the mixture was shaken under a hydrogen atmosphere (ca.30 psi) at 25 "C until hydrogen uptake ceased (60rnin). The mixture was then filtered through a Celite mat and the solvent was evaporated in vacuo, leaving the title compound (6.54g, 95%) of sufficient purity for further transformation. NMR (CDCI,): 6 7.38-7.17 (m, 5 H), 2.76 (t, J = 6.8 Hz, 2 H), 2.42 (t, J = 6.8Hz, 2 H), 1.73 (m, 4 H), 1.47 (m,2 H). Anal. (C12H1602):C, H. 6-(4-Methoxyphenyl)-hex-5-yn-l-o1 (26). 4-Bromoanisole (10g, 0.053 mol) and hex-5-yn-1-01(6.29g, 0.64 mol) were combined in triethylamine (150 mL). Tris(tripheny1phosphine)palladium dichloride (1.01 g, 0.0014 mol) and copper(1) iodide (0.112g, 0.59mmol) were added as catalysts. The mixture was heated at reflux for 2 h. The reaction was then cooled to 25 "C and filtered. Two volumes of ethyl acetate was added and the mixture was washed thoroughly with two portions of water. The mixture was then dried and evaporated. The crude product was

Gapinski et al. purified by column chromatography (Waters Prep LC-500) using a 10-25% ethyl acetate/hexane gradient elution system. Pure product-containing fractions were combined to give the title compound (1.75g, 16%). NMR (CDClJ: of 7.35 (d, J = 9.7 Hz, 2 H), 6.85 (d, J = 9.7 Hz, 2 H), 3.83 (5, 3 H), 3.75 (t, J = 7 Hz, 2 H), 2.48 (t, J = 7 Hz, 2 H), 1.75 (m, 4 H), 1.50 (br s, 1 H). IR, (CHCl,, cm-'): 3675,3019,1607,1509.MS: m / e 204 (M+). Anal. (C13H1602): C, H. Biological Methods. Binding Assay Studies: Tritiated LTB, preparations with a specific activity of 150-220 Ci/mmole and a radiochemical purity of 295% were obtained from Amersham (Arlington Heights, IL). Nonradioactive LTB4 was purchased from Biomol Research Laboratories (Philadelphia, PA). All other chemicals were commercial reagent-grade materials. Fresh human blood from 2 or 3 individuals was obtained from the Central Indiana Regional Blood Center (Indianapolis, IN) and pooled, and neutrophils were isolated by standard techniques of Ficoll-Hypaque centrifugation, dextran 70 sedimentation, and hypotonic lysis. Cell preparations were 190% neutrophils and 190% viable. The effectiveness of compounds to inhibit binding of [3H]LTB4to neutrophils was measured by using an adaptation of a radioligand-bindingassay developed by Goetzl and Goldman! The following were added to microcentifugation tubes: 10 pL DMSO containing different amounts of compound, 20 r L of radioligand (2.65nM [3H]LTB4),and 500 pL of cells suspended at a concentration of 2 X lo7 cells/mL in Hank's balanced salt solution containing 0.1% ovalbumin. The tubes were then incubated at 4 "C for 10 min. After the incubation, 300 pL of a mixture of dibutyl and dinonyl phthalate (7:2)was added, and the tubes were centrifuged for 2 min. The liquid was then decanted and the bottom tip of the tube was cut off with a razor blade and placed in a counting vial. The radioactivity bound to the cell pellet was measured by scintillation spectrometry. Nonspecific binding was determined by measuring the amount of the label bound when cells and [3H]LTB4were incubated with a >2000-fold excess of nonradioactive ligand. Appropriate corrections for nonspecific binding were made when analyzing the data. Results are expressed as percent inhibition of specific [3H]LTB4binding at the indicated concentrations. Each value is the mean of at least three replicates. The inhibitory activity of most compounds was evaluated on only one cell preparation. However an estimate of the precision of the measurements can be obtained from the inhibition observed with a reference compound, 5-[4-acetyl-5-hydroxyl-2-(2-propenyl)phenoxy]pentanenitrile, on all 102 cell preparations studied. At lod M, the mean percent inhibition and standard deviation for the reference compound were 93.9 and 3.9,respectively. At lo4 M, the corresponding values were 56.9 and 6.9. Assuming a linear correlation between percent inhibition and standard deviation, the following estimates were calculated for the precision at different percentages of inhibition: 90 A 4.2,80 f 5.0,60f 6.6,40 f 8.2,20 f 9.9,and 10 f 10.7. In a few cases where compounds were tested on more than one cell preparation, the precision of the measurementa were equal to or better than these estimates (i.e. compound 71,n = 4,95 f 0.5 at M, 60 f 2 at lo* M; compound 84,n = 4,95 f 3 at lo4 M, 88 f 3 at lo-' M, 40 f 5 at M). Registry No. 1, 2364-59-2;1-2,117424-80-3; 1-3,128578-16-5; 2, 29193-72-4;3, 16424-56-9;4, 5581-75-9;5, 81077-29-4;6, 9,128577-55-9;10, 128577-52-6;7, 128577-53-7; 8, 128577-54-8; 14, 107228-87-5; 11,54887-73-9; 12, 128577-56-0;13,89326-72-7; 128577-57-1; 15,128577-58-2;16, 128577-59-3;17,1927-71-5;18, 5851-37-6;19, 17924-66-2;20, 2430-16-2;21, 128577-60-6;22, 25, 128577-64-0; 128577-61-7;23, 128577-62-8; 24, 128577-63-9; 26,128599-33-7; 27,102831-36-7; 28,32968-32-4; 29,128577-65-1; 30,128577-66-2; 31,128577-67-3; 32,128577-68-4; 33,117424-79-0; 34,128577-69-5; 35,128577-70-8;36,128577-71-9; 37,117423-81-1; 40,128577-74-2; 41,128577-75-3; 38,128577-72-0; 39,128577-73-1; 42,128577-76-4; 43,117424-82-5; 44,128577-77-5; 45,117424-85-8; 46,128577-78-6; 47,117424-95-0; 48,117424-96-1; 49,128577-79-7; 50,128577-80-0; 51,128577-81-1; 52,128577-82-2; 53,128577-83-3; 54,128577-84-4; 55,128577-85-5; 56,128577-86-6; 57,128577-87-7; 59, 128577-888;60,117424-94-9; 61,117424-88-1; 58,117424-87-0; 62,117424-89-2; 63,128577-89-9; 64,128577-90-2; 65,128577-91-3; ( 6 ) Goldman, D. W.; Goetzl,

E.J. J . Irnmunol. 1982, 129, 1600.

J. Med. Chem. 1990,33, 2813-2817

2813

15-4; LTB4, 71160-24-2; m-EtO&CsH,COCl, 67326-20-9; OMeOC6H4(CH2)2C02Et, 70311-27-2; p-MeOCsH4Br, 104-92-7; HC=C(CH2)40H, 928-90-5; Et02C(CH2)2CH2Br,2969-81-5; PhSH, 108-98-5; PhOH, 108-95-2; Ph3P+(CH2)4C02H.Br-, 17814-85-6; PhCHO, 100-52-7; p-MeOC6H4CH0, 123-11-5; pMeSC6H4CH0, 3446-89-7; m-MeOC6H4CH0, 591-31-1; oMeOC6H4CH0,135-02-4;p-C1CsH4CH0,104-88-1; p-FC6H4CH0, 459-57-4; p-MeC6H4CH0,104-87-0; p-HOC6H4CH0, 123-08-0; 2&dimethoxybenzaldehyde, 613-45-6.

66,128577-92-4;67,128577-93-5;68,128577-946; 69,128577-95-7; 70,128577-96-8;71,117423-95-7;72,128577-97-9 73,128577-98-0; 74,128577-99-1; 75, 128~8-00-7;76, i285w-01-8; 77,117424-97-2; 78,12~578-02-979,128578-03-0;80,128578-04-1; 81,117424-99-4; 82,117425-00-0; 83,128578-05-2;84,117423-74-2;85,128578-06-3; 86,128599-34-8; 87,128578-07-4;88,128578-085; 89,128578-09-6; 90, 128578-10-9;91,128578-11-0;92,128578-12-1;93,117423-71-9; 94,117425-07-7;95,117425-08-8; 96, 117425-02-2;97, 117425-03-3; 98, 128599-35-9;99, 128578-13-2; 100, 128578-14-3; 101, 128578-

A Free-Wilson/Fujita-Ban Analysis and Prediction of the Analgesic Potency of Some 3-Hydroxy- and 3-Methoxy-N-alkylmorphinan-6-one Opioids' Zurisaddai Herntindez-Gallegos and Pedro A. Lehmann F.* Departamento de Farmacologia y Toxicologia, Centro de Investigacidn y de Estudios Avanzados, Instituto PolitQcnico Nacional, MQxico C.P. 07000, D.F., Mgxico. Received September 14, 1989 Herein we describe a Free-Wilson/Fujita-Ban QSAR (quantitative structure-activity relationship) analysis of the analgesic potency of over 50 semisynthetic opioid narcotics. The 3-hydroxy- and 3-methoxy-N-alkyhorphinan-6-ones of B/C-cis and -trans stereochemistry include compounds exhibiting structural variation at five positions [N-methyl ((217))oxygen at C3, C4-C5 oxygen bridge, alkyl substituents at C7 and C8]. The pharmacological parameter correlated was the analgesic potency (-log EDN) exhibited on abdominal contractions produced by acetylcholine injection in mice. A satisfactory correlation was obtained only by assuming interdependent contributions of the substituents on C17 and O(C3), with which it was possible to explain 75% of the variance. Phenolic compounds (3-OH) behave somewhat differently from the methyl ethers (3-OCH3), and in both series the substituents on C8 have a size-dependent negative contribution, implying steric hindrance at their contact point on the receptor. With use of this correlation the potency of five further members of the series was predicted. Subsequent testing fully confirmed the validity of the correlation since the measured potencies were, within experimental error, equal to those calculated. In a further refinement, phenolic compounds were considered separately from the ethers, and it was found that the contribution of the substituents on (217, C7, and C8 remained similar in sign and magnitude but not that of the furan oxygen. This analysis allows us to conclude that if both phenolic and nonphenolic members of this series act on the same receptor they must bind at different subsites or in alternate modes, supporting an earlier proposal in the literature.

Introduction Opiate narcotics present a Janus-like double aspect: on the one hand their abuse and its consequences is one of this century's scourges, but on the other, their use in pharmacotherapy is essential and no adequate substitutes are known. For both reasons an understanding of their mechanism(s) of action at all levels, but specially at that of their receptor(s), is one of the great challenges in pharmacology today. Biological data on narcotic analgesics2include both antinociceptive potencies in whole animals and receptorbinding assays: the correlation between them is generally good for closely related analogue^^-^ but not for very different structures.6 In addition to countless SAR (structure-activity relationship) studies over more than a century, in the last 20 years several QSAR (quantitative structure-activity relationship) analyses have also been described. Kutter et al.' showed that lipophilicity was not Taken from the M.S. Thesis in Pharmacology of Z.H.-G., C. 1.E.A.-I.P.N., 1989. Presented in part at the XI1 Congreso Nacional de Farmacologia, Pitzcuaro, Michoach, November 27-30, 1988; Abstracts p. 119. A good overview and extensive references can be found in QSAR of Analgesics, Narcotic Antagonists, and Hallucinogens; Barnett, G., Trsic, M., Willette, R. E., Eds.; NIDA Research Monograph 22, U S . Department of Health, Education, and Welfare: Washington, D.C., 1978. Wilson. R. S.: Roeers. M. E.: Pert. C. B.: Snvder. S. H. J. Med. Chem. '1975,.18,240.' Rogers, M. E.; Ong, H. H.; May, E. L. J. Med. Chem. 1975,18, 1036. Iorio, M. A,; Klee, W. A. J. Med. Chem. 1977, 20, 309. Jacobson, A. E. in ref 2, p 129. 0022-2623/90/l833-2813$02.50/0

decisive for their analgesic potency but determined their transport into and concentration in the brain. Jacobson et a1.* found it necessary to include both receptor affinity and log P terms to obtain an acceptable correlation for a structurally diverse group. Johnson in a wide ranging studyg of receptor affinities and using a Free-Wilson-related fragment approach concluded that the use of molecular hydrophobicity and steric bulk parameters did not give satisfactory correlations and emphasized the need to consider the substituents' location in the molecule. Lien et al.1° found for a series of 14-hydroxycodeinones that quadratic log P and separate molar refraction terms for substituents were necessary to give a satisfactory correlation. Katz et al." found a parabolic dependence on log P for agonists and antagonists in their affinity for receptors in the guinea pig ileum; the best correlation equation required both affinity and log P terms and was not statistically significant at the 0.95 level. In a later paper Katz et a1.,12 using the Free-Wilson/Fujita-Ban (FW/FB) method to analyze a series of benzomorphans, could explain 80% of the variance using only structural contribu(7) Kutter, E.; Herz, A.; Teschemacher, H.-J.; Hess, R. J. Med. Chem. 1970, 13, 801. (8) Jacobson, A. E.; Klee, W. A.; Dunn, W. J., I11 Eur. J. Med.

Chem. 1977, 12, 49. (9) Johnson, H. in ref 2, p 146. (10) Lien, E. J.; Tong, C. L.; Srulevitch, D. B.; Dias, C. in ref 2, p 186. (11) Katz, R.; Osborne, S.; Ionescu, F.; Andrulis, P., Jr.; Bates, R.; Beavers, W.; Chou, P. C. C.; Loew, G.; Berkowitz, D. in ref 2, p 441. (12) Katz, R.; Osborne, N. F.; Ionescu, F. J. Med. Chem. 1977,20, 1413.

0 1990 American Chemical Society